Quad Core Phasing and Information : Read it here!

[kayl]

k everyone back on topic hehehe.

So has anyone had the chance via customer to figgure out what cartridge head load/ resistor load = quadcore
that or have access to a quad core to bench them selves?

[Ssilencer]

I already talked with a friend that has a quad, he don't want to mess his mobo with insulation and stuff, but I will have access to measure power consumption with my clamp meter next week.

[Stapler]

Hold up there Ssilencer, if I'm not mistaken, TDP doesn't equal power draw. I think the most accurate method to determine quad core heat output is a direct comparison between an oc'd proc and a load tester.

[Ssilencer]

Hold up there Ssilencer, if I'm not mistaken, TDP doesn't equal power draw. I think the most accurate method to determine quad core heat output is a direct comparison between an oc'd proc and a load tester.

Check page 4, 5 and 6 please.

[Stapler]

^I read the thread as it was being posted.

I guess the way I see it, is that the overall power used is going to be greater than the heat produced by the processor. So if you take a measurement of the power consumption and tune your units to that value on a load tester, you'll end up with a slightly over charged unit. I suppose you can make up for that with an accumulator, and it might actually be benificial in the long run if future processor produce slightly more heat than current quads.

[Sneil]

So, can I take the comments I made about your last post back? You seem to have calmed down a tad. I am new to this field, and Shirley not an expert (of course I'm serious, And don't ever call me Shirley) (If you don't know the reference or the humor, sorry).
Anyway, I might check into your board to see what's up. Can you link me into any discussions of failed LS's or quad-capable units?
With respects.......Harv.

lol yea sorry I went a little hysterical on the first post there but I was a little upset. Anyway Drew is tuning his DD-X unit to 225W-300W now and will be doing some extensive testing so you all can take a look. Here is a link to the unit, make sure you go through it all and see the detail he goes into explaining everything. It'll be reopened soon with the new load test results. Anyway, carry on with the thread.
http://forums.extremeoverclocking.co...d.php?t=231428

[runmc]

lol yea sorry I went a little hysterical on the first post there but I was a little upset. Anyway Drew is tuning his DD-X unit to 225W-300W now and will be doing some extensive testing so you all can take a look. Here is a link to the unit, make sure you go through it all and see the detail he goes into explaining everything. It'll be reopened soon with the new load test results. Anyway, carry on with the thread.
http://forums.extremeoverclocking.co...d.php?t=231428

thanks for the link buddy I'll subscribe to that one

[epion2985]

Epion I don't think he's claiming that their are compressors which more more heat than the Nf9fx with less energy consumption or more quietly, simply that it doesn't move enough heat for quad cores.

Actually he was:

"Some companies have very efficient compressor ranges that have higher displacement and lower power consumption than the popular Danfoss models."

Embarco (Aspera) does have a fine line of "High Efficiency" compressors. Jinu and Johann have both used them and are very happy with them. These may be the next generation of Phase cooling compressors to handle the quads.

Can you link a data sheet that would prove this? I have looked and tried embraco compressors and wasn't horribly impressed, mostly with the noise. They also seem to be made very poorly and not neat, the shell is not regular and symmetrical and neat like danfoss.

[Gray Mole]

I just had another look at @itor's page. It's been a while since I first saw it, and it looks better than ever.

I feel kind of bad now, because maybe it seems like I was insulting Kayl's or anyone else's evaporators. That wasn't my intention, only to give Ssilencer and @itor the credit they were due. Kayl's evap is very very good, and right up there in performance with the latest Chilly1, LD's spiral, Joe Cool's stepper etc etc. It's just that the @itor evap outperforms them by a fair bit especially where high load is concerned.

It's still got limits, like anything does. I found that at 230w I started to get enough 'leftover' refrigerant that I'd be looking at an accumulator and some added heat exchange on the suction line to be able to increase capacity greatly, but it was very interesting to see how much more load I could work with so easily.

Over 200w, things become very challenging, and while some of the builders might not like to say that in public, it's very true. At higher than that, most evaps are letting a heck of a lot of refrigerant pass and we've got to do something with it before it get's back to the compressor. Even @itor's will start to show the same thing, just at a higher level of capacity.

If you want 250 or 275w of capacity you've got to take that into consideration and plan your pipework accordingly.

In transport and commercial refrigeration you don't find this kind of issue, but the evap can be much larger, and you're able to plan things so that overcharging isn't much of an issue. The system size is much larger, so there's 'room' for the refrigerant to be vapor well before it get's anywhere near the compressor.

We're dealing with a tiny contact patch and little more 'supporting' surface area in the body of an evap. That means we've generally got to try to flood it as heavily as we can to get enough refrigerant to contact that spot and then we've got to 'clean up the mess' afterwards by trying to evaporate it before it gets to the compressor and breaks it.

That's all got to be done in an incredibly small space that'll still fit into a 'standard' size phase case. More pipework isn't really an option anymore, so we've got to get smarter.

Wrapping the captube in our systems enhances subcooling, yes. It also turns your suction line into a slhx, and while that's been working ok up until now, it's starting to become inadequate for higher loads.

Better evaps are going to help, but even @itor's evap has it's limitations.

So increasing the heat exchange at the suction line is the only way to deal with it.

Improving the rest of our systems, the condensor, the captube methods, they will all make a difference in capacity. But when it comes to the end, the evap is going to take out the heat it can, and we're going to be left with a bunch of refrigerant in liquid form that's going to have to be dealt with.

The compressor can really help in getting better temps at certain capacties but as it's been said before, a compressor doesn't make the capacity happen, it just sucks and blows. A compressor has a given capacity at a certain temperature with a specific gas. A compressor's capacity isn't set in stone, as it's just a relationship.

I only see 2 choices for increasing capacity greatly in our systems.

1. increase the suction volume to a level allowing room for vaporisation.

2. increase the level of heat exchange on the suction line to a level that allows for the needed increase in charging.

Using a better evap helps a lot if you can live with it's level of efficiency, sure. If you want more than what it wants to give, however, then you've got to change your system so that you're taking into account the refrigerant 'waste' and cleaning it up before it kills the compressor.

Hope that all made sense.

Gray

[Gray Mole]

Oh, and yeah the NEK series of compressor is very good. I'll have to try them in their larger models, as i've only used the 6-8cc ones. Still quite good, and very quiet.

I've had the pleasure of seeing some of the more experimental compressors available but I had to sign something saying I couldn't talk about it

The noise levels and performance that's in the future is pretty impressive though, but I have a feeling that the ones I got to see are going to be out of our price range for a good long while

I'll always love Danfoss' NL series though. I just wish they weren't as expensive as they are. I've had more powerful compressors, and quieter compressors in similar cc's, but nothing with the combination of quiet AND powerful that these have.

Cheers

Gray

[runmc]

Can you link a data sheet that would prove this? I have looked and tried embraco compressors and wasn't horribly impressed, mostly with the noise. They also seem to be made very poorly and not neat, the shell is not regular and symmetrical and neat like danfoss.

Well actually I'm not trying to prove anything. I just know what I've been told from two of my trusted friends. Here is a link to a very new line of Embraco NEK compressors

I would really like to try these compressors myself. I will be as soon as I get one ordered.

[epion2985]

Well actually I'm not trying to prove anything. I just know what I've been told from two of my trusted friends. Here is a link to a very new line of Embraco NEK compressors

I would really like to try these compressors myself. I will be as soon as I get one ordered.

Thanks, I was just very curious, thanks for the link.

[harvshark]

lol yea sorry I went a little hysterical on the first post there but I was a little upset. Anyway Drew is tuning his DD-X unit to 225W-300W now and will be doing some extensive testing so you all can take a look. Here is a link to the unit, make sure you go through it all and see the detail he goes into explaining everything. It'll be reopened soon with the new load test results. Anyway, carry on with the thread.
http://forums.extremeoverclocking.co...d.php?t=231428

Thank You for the apology and the link. There is nothing wrong with constructive dissent. That's called debate. That's what this forum is about. I'll peruse the link. Thanx again. 1HP rotary??? Cool! (No pun intended.)

[johann]

Since I used NEK 2150GK I will never use a Danfoss NL11F if I have the choice. Simple as that, its far superior in all ways

[hipro5]

Since I used NEK 2150GK I will never use a Danfoss NL11F if I have the choice. Simple as that, its far superior in all ways

I've used them both as single stageres and as second stageres on cascades....NL11F simply can't be touched.....

[epion2985]

I've used them both as single stageres and as second stageres on cascades....NL11F simply can't be touched.....

_

[harvshark]

lol yea sorry I went a little hysterical on the first post there but I was a little upset. Anyway Drew is tuning his DD-X unit to 225W-300W now and will be doing some extensive testing so you all can take a look. Here is a link to the unit, make sure you go through it all and see the detail he goes into explaining everything. It'll be reopened soon with the new load test results. Anyway, carry on with the thread.
http://forums.extremeoverclocking.co...d.php?t=231428

Had the time this morning to view the thread in the link you sent. Very nice, lots of great pics. Drew's work looks top notch. I thank you again for the link, I registered on eocf and I am sure I will stop by once in a while to see what's going on there. Even if I do more lurking than posting.

[Sneil]

Great sounds good glad to have ya, make sure you do post every once in a while

[epion2985]

NL is quite an old design, NEK much newer. The Danfoss compressors are exceptionally durable and well proven technology, not having used them i can't say for sure but i believe Johann...

Tom

Then you are saying hipro5 doesn't know what he is talking about?

Two contradicting opinions here.

[Gray Mole]

Hey now, no need to try to start a fight

Hipro's opinion is shared by many around here. The NL11F and the NF series for 115v is well known as being the best around for it's combination of low noise and high power.

I've always had good success with them myself.

The NEK series does look good though, and I'm looking forward to testing the larger CC versions at some point.

Technology moves forward, always does. The fact that the Danfoss compressors have been 'king of the hill' this long really says something about how good they are. One day, though, like anything, something will come along and take the crown away.

Will it be the NEK Aspera/Embraco compressors? I don't know, but whether it's that or another type of compressor it's got to happen eventually

I've already posted about my feelings that a compressor only dictates temperature at a given load, and not the overall system capacity itself, so it's not even really 'on topic' because for the most part you can still use a 1/5HP compressor and tune for 250+ watts, you'll just get a higher temp than a larger compressor.

Cheers

Gray

[hipro5]

Talking about Watts.......
One of the old Chilly's evaps which is made for VGAs - it has on it's side the suck and it's captube - can handle about 350Watts at -82.7*C on an esdee's cascade at full load at ALL the time......This is the max temp drop it has.....On the most 3D, it works -87*C load.......
ALSO as I said some pages back, elevens evap can handle a Kentsfield at 3850MHz and more than 350Watts all the time......
Now eleven made for us - as we told him - a 5cm evep instead of a 4cm, so as to test it too....

[Ssilencer]

Talking about Watts.......
One of the old Chilly's evaps which is made for VGAs - it has on it's side the suck and it's captube - can handle about 350Watts at -82.7*C on an esdee's cascade at full load at ALL the time......This is the max temp drop it has.....On the most 3D, it works -87*C load.......
ALSO as I said some pages back, elevens evap can handle a Kentsfield at 3850MHz and more than 350Watts all the time......
Now eleven made for us - as we told him - a 5cm evep instead of a 4cm, so as to test it too....

Nice to see you around hipro5
You talk about a chilly evap handling high load on a CASCADE and what was discussed until now was single stages, there are no much info posted here (this thread) on cascades, just SS. And the experience on SS is more than 100 to 1 in favor of SS. BTW, that was a load tester at 350watts?

Nobody told that any evap can't handle high loads, but how efficient handles it.
You can do an evap of 10 cm tall, it is not a matter of height, it is a matter of internal surface and design.


As always sory for my bad English :P

[hipro5]

I'm talking about a 5cm base and not height.....All of nowadays evaps have 4cm base....

No heatload, only a 8800GTX at 1.82VCore and 918MHz/1215MHz....Do you think it's less than 350Watts?.....

Talking about cascades: I have seen one thing....IF a cascade can "handle" without "going down (up)" a temp, a single stage can do it too....R1150 is a "soft" gas....R410 is a "hard" gas and R507 is a "midlle" gas.....
A SS could handle a QUAD at a certain temp.....
Mine - moded Prommie - can handle about 250Watts at -46*C evaps BASE temp load......ALL the time.....

Talking about cascades: How many could handle a QUAD at 3850MHz?....Not LN2.....A cascade at -82*C full load.....

My idea is to build an evap like eleven's which "flows" ALL the gas on the base....and not from up to down and visa-versa.....ALL the gas ON the base.....

[Gray Mole]

The design characteristics of a cascade normally make higher loads FAR easier to tune for.

It's the slhx setups and generally having more suction volume that makes it easier to tune high load into a cascade.

That's part of what I'm trying to adapt into SS coolers, some of the cascade techniques not for low temp, but for high capacity.

When it comes down to it, I wish everyone would try an @itor evap, and just tune it into a system the same way as they would any other high end evap.

I think you would be able to see what Johann and I got to see with respect to the pressures, load characteristics, and general performance of @itor's evap.

I'll have to try Eleven's Evap as well, it does sound very good too.

When it comes to anything over 250w though, it doesn't matter what head you use I think, looking at upgrading the suction pipework, and general spec of the units, is the only way that high loads are going to be dealt with correctly, without massive floodback issues.

Cheers

Gray

[esdee]

Nobody told that any evap can't handle high loads, but how efficient handles it.
You can do an evap of 10 cm tall, it is not a matter of height, it is a matter of internal surface and design.

As always sory for my bad English :P

actually hipro5 has to be more sorry for his english!
sslinencer you are right about the cascade Vs sstages...

as for the evaps design i have many ideas the last couple of months...
1) i don't think base thickness must be more than 5mm. i have the feeling that with large bases we try to remove heat from a point too far from the cpu core itself!
2) mass mass mass... i think mass is important in a weird way.
an evap with more mass has more heat buffer to deal with the load until the systems low side responses.
for example -100*C idle, apply 200W --> instant drop to -89*C then system stabilizes at -93*C

but when overclocking, if i start my 32m at the limit MHz for -93 i will get an error when the phase instantly drops to -89, so i have to bench in the Mhz where i am stable at -89*C! i need to chop that delta between instant drop temp and i think this is how mass helps high loads

3) so we need mass, but not base thickness, so what we did is increased the diameter

(ssilencer that was what hipro5 was talking about when saying 'increased 4cm to 5cm' )







this new baby from our friend eleven will be tested next week on a cascade that we know how it has been performing with the same design/base thickness 40mm version! (hope we find something )